Research

This page provides an overview of my research conducted during my PhD and my time at the OECD. Some of these papers also build on work from my master’s thesis, which focused on Bayesian Neural Networks.

Working Papers

Transmission Channel Analysis in Dynamic Models

2025 (first edition 2024)

We propose a framework for analysing transmission channels in a large class of dynamic models. We formulate our approach both using graph theory and potential outcomes, which we show to be equivalent. Our method, labelled Transmission Channel Analysis (TCA), allows for the decomposition of total effects captured by impulse response functions into the effects flowing through transmission channels, thereby providing a quantitative assessment of the strength of various well-defined channels. We establish that this requires no additional identification assumptions beyond the identification of the structural shock whose effects the researcher wants to decompose. Additionally, we prove that impulse response functions are sufficient statistics for the computation of transmission effects. We demonstrate the empirical relevance of TCA for policy evaluation by decomposing the effects of policy shocks arising from a variety of popular macroeconomic models.

Publications

Quantifying Uncertainty of Portfolios using Bayesian Neural Networks

International Joint Conference on Neural Networks (IJCNN)
2024

Quantifying the uncertainty of a financial portfolio is important for investors and regulatory agencies. Reporting such uncertainty accurately is challenging due to time-dependent market dynamics, non-linearities in the return and risk properties of a portfolio, and due to the unobserved nature of the market risk. We propose Bayesian Neural Network (BNN) models, namely Recurrent Neural Network (RNN) and Long Short Term Memory (LSTM) models, to estimate the time-varying return distribution of an asset portfolio. The proposed models estimate the density of returns and incorporate parameter uncertainty through Bayesian inference. The uncertainty and any financial risk metric of interest can directly be obtained from the estimated density. Furthermore, through the BNN input-output design, proposed BNNs incorporate potential non-linear effects of each asset in the portfolio on the obtained density estimates. The proposed method is applicable to assess the uncertainty of any portfolio where the portfolio weight optimization is separated from risk assessment. We analyze the risk of a daily, equally weighted portfolio of 29 ETFs and a risk-free asset for a long time span with differing market environments between 09/06/2005 and 10/09/2020. We study the effects of different inference methods on the obtained results. The proposed models improve portfolio risk estimates compared to the benchmark. The performances of the proposed models depend on BNN design and the inference method. RNN models lead to relatively more stable results compared to LSTMs. Furthermore, the results of models with a relatively higher number of parameters depend heavily on the estimation method.

OECD Working Papers

Using Unit Value Indices as Proxies for International Merchandise Trade Prices

2022

In light of the need for detailed and timely internationally comparable trade price indices, this paper describes a multi-tiered methodology to mitigate many of the empirical challenges associated with using customs data, to provide more robust estimates of unit value indices (UVIs) by country and product. UVIs are available for both exports and imports, by reporting country and the CPA 2-digit level of classification. Although the approach cannot capture changes in the quality of products nor compositional changes happening at a lower than HS 6-digit classification, the results indicate that at higher levels of aggregation (SITC 1-digit level), estimated UVIs closely follow price changes obtained from other sources. This is observed both for products with significant and rapid quality changes, such as hi-tech products, and for products with a low rate of quality changes, such as commodities, other primary and low-tech goods. Furthermore, products where little quality change occurs over time show similarity between UVIs and price changes from other sources at lower levels of disaggregation. The methodology is used to produce the Merchandise Trade Price Index and the data is made publically available on .Stat under the International Trade and Balance of Payments heading.